Turkey’s news media landscape in Twitter: Mapping interconnections among diversity

News media network: A relational approach

The following part of this article concentrates on how the above-mentioned media accounts are connected in Twitter and investigates the nodes (representing the media accounts) and edges (links between those accounts) to expose which clusters dominate the news media landscape. Connections between accounts are established based on a variety of factors such as the type of organization (news agency, alternative media, etc.), ideologies, media ownership, institutional policies on social media (can/cannot follow), issue-specific focus (gender, minorities, etc.), representation (or self-definition), and relations with news sources. In order to illustrate those connections within an intermedia setting, indegrees (followers) and outdegrees (followings) are investigated in two directed graphs, while the media categories are listed.

Figure 2 demonstrates a map of Turkey’s media landscape in Twitter with regard to the indegree of each media account. Several nodes from diverse media categories have higher indegree, meaning they are more apparent in the network. Dicle Haber Ajansı (19), mostly attracting attention from the activist media, has the highest indegree. Bianet (16), one of the earliest civic media in Turkey, ranks second having connections from its own category, as well as from activist media and women’s media. BBC Türkçe, Sendika.org, Agos, T24, and İMC Tv are among the other nodes having higher indegree. These media, to a great extent, are followed by the same set of nodes, which include certain accounts of activist media, dissident institutional media, and women’s media. News agencies, in general, have high indegree but are very different in character. For instance, DHA (13) primarily connects to the mainstream media, while other prominent news agencies that typically cover issues on Kurdish minorities such as Fırat Haber Ajansı (13) and ETHA (11) have connections both from ethnic and alternative media.

OPEN IN VIEWER

Figure 2.

Turkey’s news media network in Twitter. Node size proportional to indegree.

According to Boyd and Ellison (2007: 211), participants are not necessarily ‘networking’ or looking to meet new connections on social media; instead, they are primarily communicating with the ones that are already a part of their extended social network. Their argument may indicate the reasons behind the loose connections of mainstream media in Twitter. In this study, for instance, it is not a coincidence to see Radikal, a mainstream media with left-wing orientation, to be followed by a variety of nonmainstream media accounts. However, pro-government media remain relatively low in the number of their indegree connections as there is a number of nonmainstream organizations in the network that oppose the government. Mainstream media generally have low indegree except for CNN Türk, Hürriyet, Milliyet, and NTV, which did not present a distinct political tendency in the past.

In terms of outdegree connections, two related functions are evident: monitoring other news media and using Twitter as the source of information. Monitoring usually helps particular news media organizations with limited workforce and financial sources to keep track of the agenda. Those organizations generally reproduce the content that is already available in Twitter, therefore amplify certain issues of their interest. Twitter is also a unique source of information for news organizations. It provides journalists with the ability to identify and reveal emerging issues. This does not only enable journalists to cover more issues but also helps enrich their stories.

Figure 3 clearly shows that these functions differently apply to the media categories. For instance, the mainstream media do not follow any accounts other than the ones affiliated with their ownership group, whereas the nonmainstream media are typically dispersed around the network. In total, four media categories, namely, dissident institutional media, activist media, civic media, and women’s media are noticeable regarding their outdegree connections. However, not all the nodes in these categories share the same tendency. The dissident institutional media organizations, such as Yarın (50), Demokrat Haber (39), and Gerçek Gündem (34), have outdegree connections to a variety of accounts including the mainstream media. The activist media accounts mostly have links to others in their own category. In the civic media category, Jiyan (40) and P24 (30) are noticeable because they link to multiple accounts from diverse categories. The women’s media have outdegree connections to certain accounts, which from time to time cover gender issues.

OPEN IN VIEWER

Figure 3.

Turkey’s news media network in Twitter. Node size proportional to outdegree.

The Internet redefines the relative position of journalists and audiences and opens a channel for a dialogical kind of journalism through which news production becomes a collective endeavor (Mitchelstein and Boczkowski, 2009: 573). Twitter is a substantial platform for the nonmainstream news organizations helping them connect to their peers, as well as reach wider audiences almost without cost. Its feeds combine news updates of 140 characters or less, resulting in the collaborative construction of events out of atomized stories and stories out of subjectively experienced events (Papacharissi and de Fatima Oliveira, 2012: 267). Considering the repressed and the minorities, the ubiquitous nature of Twitter allows the nonmainstream news organizations to collaborate and cover issues that are often ignored by the mainstream media. Therefore, the linkages they establish in Twitter are fundamental for their daily operations. In this setting, it is not surprising to see the activist media accounts to have high outdegree, as many of them are run by volunteering citizens supported by media professionals. Twitter itself is the source of news for the nonmainstream media.

Both maps presented in Figures 2 and 3 portray a mise en scène where connections indicate a dialogic form of journalism (Deuze, 2003). To put it briefly, media accounts of the same category have more intensive connections compared to intermedia connections, although the ties among the mainstream media are apparently scarce. Affiliations are quite visible in their connections as they tend to follow particular media organizations owned by the same proprietor. For instance, Sabah Gazetesi, A Haber, Daily Sabah, and Takvim together form a cluster representing the Turkuvaz Media Group. ⁶ However, the dissident institutional media follow high number of accounts, whereas the news agencies do not follow other media but have high indegree as the providers of news. There is a tendency among the nonmainstream media accounts to connect regardless of media categories. We need to underline the fact that many of those media organizations neither circulate print copies nor broadcast through cable or satellite services, and not all of them necessarily host a website. While the civic media are available online through their websites, the activist media in particular can only be reached through the SNSs.

The patterns of news circulation and impressions on gatekeeping

SNSs are primarily organized around people, not interests. They are structured as personal (or ‘egocentric’) networks, with the individual at the center of their own community (Boyd and Ellison, 2007: 219). Research on social media indicates that Twitter has a limited role in providing news about the interests of specific communities, while Facebook is used by members who wish to stay connected with people they know offline (Pew Research Center, 2015). Facebook users, in that sense, tend to get news on their social environment through their existing social networks. Compared to Facebook, Twitter has a distinct configuration allowing an exclusive focus on issues, but it proposes a similar scene through filter bubbles. In Twitter, users prefer to see posts they ideologically agree with, even including the seductive falsehoods (Thompson, 2017). Particularly on explicit political issues, individuals are clearly more likely to pass on information they receive from ideologically similar sources rather than to pass on information they receive from dissimilar sources (Barberá et al., 2015: 1540).

As the findings from this research indicate, there are clusters in Twitter, each representing a news pool. These pools serve as information ghettos (Waltz, 2005: 21), where audiences get connected to their preferred media organization to access information based on the subject of their interest or political tendency. This helps media organizations sustain their gatekeeping role in Twitter similar to the conventional media platforms. From this point of view, journalists have the authority to make choices about what to tweet and to decide on the content that is shared (Hermida, 2010b: 300). Mainstream news accounts in Turkey have far more followers than the remaining accounts; therefore, they can be considered as the leading candidates to serve the foundations of gatekeeping in Twitter. In order to better illustrate this argument, we should interpret the findings from Figures 2 and 3 together with the numbers presented in Table 1. In contrast to network density, a greater number of followers are present in the portions of the network, where the nodes are loosely connected. In other words, there is a negative correlation between network density and the number of followers. The positioning of the mainstream organizations within the news environment in Twitter presents evidence on their role as gatekeepers.

Research on social media also underlines the importance of indirect exposure from user networks. Compared to their audience through direct subscription, media organizations reach considerably a larger audience through indirect exposure via social links, such as through a friend who follows a particular media source or via a retweet from a friend (An et al., 2011). For instance, even though you are only following your local, regional newspaper in Twitter, and not following the New York Times, people whom you follow are more likely to recommend to you an article from the New York Times compared to an article from your local newspaper. In this case, even though you are not directly following the New York Times, you are likely to click on the New York Times article (Hong, 2012: 73). In this approach, indirect exposure from user networks steers the sort of content available to the audiences on social media and results in the promotion of certain sharings.

Although to a lesser extent, the above-mentioned cycle works the other way around to help disseminate the nonmainstream content in Twitter. This leads to another approach to understand news diffusion in Twitter. As we are witnessing the rise of nonmarket, distributed, and collaborative forms of journalism practiced in the platforms for political mobilization and organization (Benkler, 2007: 465), we need to rethink the concepts of gatekeeping and agenda setting and bear in mind that the environment in which such concepts are elaborated is under constant change. An infinite number of people simultaneously serve as sources, audiences, and information providers in this environment, where journalists no longer have much if any control over what they will see, read, or hear (Singer, 2007: 82). From this point of view, Twitter can be considered an equal opportunity platform for the members of small-scale media to present their content. Inclusion of small-scale media to the news sphere in Twitter also triggers an increase in the amount of content available to audiences and provides them with the agency to filter and spread the news of their own choice.

The fundamental parameter for reviewing the gatekeeping theory with respect to social media is the presence of connections between the media and their readers. Social links broaden the types of information users receive and allow less popular media to gain more presence (An et al., 2011). Accordingly, it is likely for a reader following certain nonmainstream accounts to receive news from other nonmainstream sources that are not directly connected to the reader’s account. For instance, there is a good chance for a reader who follows Kamera Sokak to see any content shared by Naber Medya TV. As pointed out in this research, there is higher density in the part of the network, where the nonmainstream media are positioned. This notional space, predominantly occupied by civic and activist media, clearly demonstrates how gatekeeping works in Twitter and at the same time coincides with the rhizomatic nature of alternative media. However, this is further challenging due to the fact that the number of followers owned by each media account differs widely.

Another fact to be highlighted is Twitter’s amplifying effect. Twitter is a platform through which content that is already available on the Internet is propagated. However, this function very much depends on the density of the network and relies on the number of sharing account’s followers. The literature indicates that mass media in Twitter can reach a large audience but do not follow others actively (Cha et al., 2012). Similarly, the mainstream media accounts in Turkey have high number of followers; however, the number of their connections is not as much. On the contrary, the nonmainstream media accounts have less number of followers but have more connections. This means that the content shared by the mainstream media accounts may reach larger audiences, whereas the sharings of the nonmainstream accounts are particularly available to the audiences of a closed community. It must also be noted that the type and originality of the shared content considerably affect its propagation.

Remarks on media polarization

The literature on social media involves two primary approaches considering the aspects of polarization. The echo chambers approach focuses on the highly fragmented, customized, and niche-oriented aspects of social media and suggests that these venues foster greater political polarization of public opinion. However, the crosscutting interactions approach argues that the openness of the Internet and social media allows citizens to receive a wide spectrum of ideologically different views (Hong and Kim, 2016). Indications of polarization in Twitter can be reviewed with respect to these two approaches, while considering several factors such as partisanship, political stance of media owners, and definition of organizational roles.

According to the echo chambers approach, news media in Twitter contribute to the polarization of public opinion. This argument corresponds to the normalization thesis which suggests that the Internet increasingly reflects the political forces that dominate politics and social life in the real world (Margolis and Resnick, 2000: 3). From this perspective, Twitter can be seen as a platform to reiterate the conventional ideologies and routines. A recent report of Genel İzleyici explores the closeness of connections among the mainstream media outlets in Turkey based on the audiences they have in Twitter (Mat, 2016) and suggests a grouping of users who tend to follow the same media. ⁷ Another study, conducted by Irak (2016), provides an analysis of the Twitter interactions regarding the AA board members and points out the issue of political concentration in Turkey’s public service media:

AA, along with the TRT [the national public service broadcaster], has always presented a pro-government editorial line and political clientelism; with its 90 percent of content being devoted to the AKP, board members being former consultants of Arınç and Erdoğan or columnists of the ‘pool media’, and the cozy relationship founded in Twitter with other influential ‘New Turkey’ supporters do not depict a healthy picture for the Turkish media or the country’s democracy. (p. 20)

The fact that many mainstream media institutions in Turkey are affiliated with the ruling party, their relations with other institutions maintain under certain limitations. The mainstream media in Turkey lack the freedom and editorial independence simply because 85 percent of those media outlets are controlled by proprietors, who are for the most part engaged in vast businesses other than media (Baydar, 2013). Dependency on the government for economic interests results in press-party parallelism (Seymour-Ure, 1974). According to Blumler and Gurevitch (1995: 65), the highest degree of partisan involvement occurs when the parties are directly associated with the running of media enterprises via ownership, as it is the case for many media outlets in Turkey.

Mainstream media institutions in Turkey reflect a tendency in Twitter similar to which they maintain in their offline comfort zone. Indications of this tendency can be clearly seen in their Twitter connections, which are clearly formed around their ownership structures (Figure 4). They do not link to actors other than news agencies and manifest an online echo chamber of mass-mediated political views (Singer, 2005: 192). The nonmainstream accounts, however, maintain close ties with each other as a consequence of their dissident character and oppositional attitude toward the ruling AKP government. However, this sort of clustering (as illustrated with a dotted circle in Figure 4) is not ideal because it builds up an echo chamber, reflecting the vulnerable potential of interactive media to enhance public communications and the risk of being submerged or marginalized (Blumler and Gurevitch, 2001: 9).

OPEN IN VIEWER

Figure 4.

Illustrating indications of polarization.

The crosscutting interactions approach suggests a more democratic stand, taking audiences as active stakeholders of the media industry. In this rather optimistic view, the quality of the informational environment may increase as the audiences receive news and information from diverse ideological sources (Mutz, 2006). According to Benkler (2007: 465), we are seeing the rise of a collaborative fashion, which allows individuals engaged in public discourse to decide whom to trust and whose words to question. This results in a shift from individualistic, top-down mono-media journalism to team-based, participatory forms of journalism (Deuze, 2005: 452). Considering that many of the nonmainstream media are founded by civic initiatives, this approach helps us expand our understanding, including the further dynamics of polarization such as mobilization and activism. As Rheingold (2000) asserts, ‘the political significance of computer-mediated communication lies in its capacity to challenge the existing political hierarchy’s monopoly on powerful communications media, and perhaps thus revitalize citizen-based democracy’ (p. 14).

Journalism practiced through social media is the modus operandi for citizens, who demand to participate in the politics and culture of a society. According to Dahlgren (1996: 62), journalism in this mode serves as an integrative force and as a common forum for debate. For him, the fundamental role of journalism in democracy is to link citizens to political life because the problems derive basically from the character of the political life and power relations that in a variety of ways hinder participation (Dahlgren, 2009: 48). Twitter, in this setting, functions as a platform for the civic media and provides them with the ability to reach the audiences and challenge mainstream ideologies. Therefore, the aggregate linkages of the nonmainstream media can be conceived to represent the civic network of dissident voices.

Lorentzen’s (2014) analysis on communication and relationships across political boundaries also reveals clear examples of polarization but further suggests that people are using the opportunities of Twitter to create connections across boundaries. The notion of connectedness is an indispensable ideal particularly for the nonmainstream organizations as not only do they reach their audiences through their linkages, but they also form communities while incorporating them in the news production processes. In their research on group polarization in Twitter, Yardi and Boyd (2010) found that communication between like-minded individuals strengthened group identity, whereas communication between different minded individuals reinforced in-group and out-group affiliation. A similar phenomenon can be observed in this research, which reveals significant homophily within media categories. However, even if the presence of homophily clearly leads to polarization, this does not necessarily mean that the clusters representing different media groups are impenetratable.

Turkey’s news media landscape in Twitter is highly polarized. The political disposition of the media organizations reflects on their positioning in the network. For instance, a breakdown of the map reveals the following ideologies from right to left, respectively: pro-AKP, liberal, leftist, and pro-Kurdish. Nevertheless, interactions in Twitter allow stronger networking, so there is a high probability for transitions to occur between different media groups. For instance, in particular cases, the nonmainstream media bring together different segments of the society around common issues and provide multifold aspects to the coverage of these issues, which can be transferred to the mainstream media agenda (Doğu, 2014, 2015). If any issue emerging within the echo chamber of the nonmainstream media can be transferred to the mainstream media domain, then polarization should not be taken as a rigid concept limiting the flow of information in Twitter. However, governmental policies and political agenda are likely to be more effective in terms of supporting or mitigating polarization.

Conclusion

In this study, the contemporary media environment is taken as a complex and networked system of social awareness (Papacharissi and de Fatima Oliveira, 2012), and Twitter is defined as a stage in this environment, where media organizations get connected in novel ways. The relational approach employed in this research helped illustrate the connections between the prominent actors in Turkey’s media field, as well as the dispositions of the nonmainstream organizations in Twitter. It has been found that Twitter’s asymmetric model facilitates the connections between media organizations in several ways, and this results in the composition of clusters mainly based on intermedia differences.

The idea of connectedness may inspire better collaboration among different actors; however, it also challenges the existing tenets of the news environment. This study suggests evidence on how connections define the diffusion of news in Twitter and contribute to the establishment of echo chambers. The social graph of news media in Turkey portrays a dynamic environment, where the relationship between various actors is redefined. The mainstream media are still dominant in terms of the number of their readers; however, the domain of nonmainstream media is also promising. Particularly, the media accounts opened in the aftermath of the Gezi protests are worthy of consideration for their connections, which were established mainly as a result of tactical operations.

This study provides a framework for future research on the implementation of journalistic practices in Twitter. It also offers valuable data for researching specific cases based on intermedia differences and polarization, while allowing us to see the broader picture in Turkey’s media landscape. However, there is much work to be done as this study is limited to the news media accounts in Turkey. Further research can be conducted to include other countries, as well as relevant stakeholders in the media industry, such as personal accounts of journalists and political actors.